Although primary pleural tumors have been reported since the eighteenth century, the epidemiology of mesothelioma first came to light in 1960 with the report by Wagner et al.1 of 33 asbestos mine workers from South Africa who developed mesothelioma. Pleural mesothelioma previously was classified as benign or malignant. However, recognition that “benign” or “localized” mesothelioma has a biology that is distinct from MPM led to a change in nomenclature. These benign tumors are now termed solitary fibrous tumors of the pleura.
MPM is a rare tumor. Although the geographic distribution of the disease is diverse, taken as a whole, the United States has an incidence just under 1 per 100,000 persons.2 The incidence has been rising since the 1970s. The male-to-female ratio is 5:1, which is likely reflective of occupational exposure to asbestos.
The clinical presentation of MPM is usually insidious. The most common presenting symptoms are dyspnea and chest pain.
Staging in MPM, as is the case in other aspects of the disease, lacks consensus. Various staging systems exist. The classic system described by Butchart et al. in 1976 is relatively simple and descriptive.3 The Brigham staging system is based on resectability by extrapleural pneumonectomy (EPP) and may not be of value in patients undergoing P/D.4 The tumor, node, metastasis (TNM) staging system proposed by the International Mesothelioma Interest Group (IMIG) is the accepted American Joint Commission on Cancer staging system.5
In the days before effective systemic therapy, MPM was thought to be uniformly fatal. Surgery was reserved for diagnosis and palliation. In the first reports of “curative” surgery, Butchart et al.3 performed EPP with a surgical mortality rate of 30%. In the nearly 30 years since the initial report, advances in patient selection, as well as intra- and postoperative management, have decreased the mortality of the operation substantially, as reported by centers with high volumes of mesothelioma surgery. Sugarbaker et al.6 reported their mortality rate from 183 consecutive EPPs performed at the Brigham and Women's Hospital as 3.8%. At Memorial Sloan-Kettering, Flores reported a 5.2% mortality for EPP.7 The staggeringly high mortality rate seen in early attempts at EPP led to a movement away from this operation and toward P/D as a method of debulking tumor. The mortality of P/D is reported to be 1.8%,8 and the lack of evidence demonstrating superiority of EPP over P/D is thought to be a consequence of the upfront mortality increase with the more extensive operation.
There are those who still believe that surgical intervention for purposes other than palliation in mesothelioma is not indicated. Although it is true that there are no randomized controlled trials comparing surgical treatment with supportive care (or other treatments) in these patients, the reality is that these trials likely will never be performed. For those who treat this disease and have a less nihilistic outlook, surgery forms a key component of the treatment algorithm.
There has been a move in the international mesothelioma community to standardize the terminology associated with the disparate operations performed by various surgeons under the umbrella heading of P/D. This should enable improved analysis of results across various centers and enable more coordinated research efforts, as to date, the majority of reports and studies are single center in origin. Recommendations published jointly by the International Association for the Study of Lung Cancer (IASLC) and the IMIG following a survey suggested that the term P/D be used to denote attempted complete removal of macroscopic tumor from the visceral and parietal pleura and the term extended P/D be used if the diaphragm and pericardium are removed as well.9
Indications for P/D can be regarded as patient-related or tumor-related. Perhaps the least controversial statement one can make about P/D is that it can be offered to patients who do not have the cardiopulmonary reserve to tolerate pneumonectomy. For patients who can tolerate pneumonectomy, the choice of operation becomes less clear. Some centers perform P/D for patients with early-stage disease, that is, confined to the parietal pleural “capsule” (Butchart I, IMIG T1a or T1b), with the rationale that if no lung parenchyma is involved, the inherent morbidity and mortality risk of adding a pneumonectomy is not warranted. Others disagree, based on the rationale that the absence of lung parenchyma facilitates the administration of postoperative adjuvant radiotherapy.
If one accepts that MPM is a disease where true R0 resections are a theoretical achievement, then the goal of surgery is to remove all gross tumors and serve as a springboard for adjuvant therapy. The choice of operation then is made based on the extent of resection required and the extent of resection the patient can tolerate. Some clinicians believe that with newer methods of radiation administration and ongoing attempts at other local and systemic therapies, the argument that residual lung parenchyma hinders appropriate adjuvant therapy may be less of a factor than it once was.10
We reported a retrospective study of 663 patients (385 had EPP and 278 had P/D) which demonstrated a hazard ratio of 1.4 for EPP when controlling for stage, histology, gender, and multimodality therapy. Clearly the study is subject to selection bias, but this reinforces our view that P/D may be the optimal procedure if adequate debulking can be performed by leaving the lung parenchyma in place.11
All patients undergoing consideration for P/D need thorough imaging and cardiopulmonary evaluation. At a minimum, pulmonary function testing should be performed. Quantitative ventilation/perfusion scans may also be indicated if associated lung resections are anticipated or to evaluate the possibility of EPP. Computed tomography (CT) scanning of the thorax and upper abdomen is required, and magnetic resonance imaging (MRI) may be superior in assessing discrete focuses of chest wall invasion or diaphragmatic muscle involvement.12 However, rarely does MRI change surgical management. Fluorodeoxyglucose positron emission tomography (PET) scanning in MPM can be used to provide stage and prognostic information. In addition to helping to determine the extent of tumor, PET scanning can be used to detect N3 or M1 disease in 10% of patients.13,14 The standardized uptake value also can be used to predict the presence of N2 lymphatic spread.14 High standardized uptake value also has been shown to correlate with poor survival in MPM.15
Another controversial question in the preoperative evaluation of patients is the role of mediastinoscopy in MPM. It is useful in determining the N stage of most patients and is more accurate than CT scanning.16 However, up to 25% of patients have lymph node involvement confined to areas of the hemithorax inaccessible by mediastinoscopy, such as the peridiaphragmatic and internal mammary regions.7 Furthermore, although N2 disease does negatively impact survival, some feel that it should not be used as the sole reason to deny surgery.
After the induction of general anesthesia, a double-lumen endotracheal tube should be inserted to facilitate the operation. An arterial line and central venous pressure monitoring are important because blood loss is often significant (approximately 1–2 L). The patient is placed in the lateral decubitus position, and a long S-shaped posterolateral thoracotomy incision extending downward to the costal margin is made (Fig. 121-1). The sixth rib is resected, and the dissection is begun in the plane between the endothoracic fascia and the parietal pleura (Fig. 121-2). The pleural tumor is bluntly dissected away from the chest wall. The plane is then developed in a cephalad direction toward the apex from the posterolateral direction. Care in identifying the subclavian vessels is prudent because a traction injury to these structures is difficult to repair (Fig. 121-3). As each area of dissection is completed, packs are placed to aid in hemostasis because a fair amount of blood loss will result from the blunt dissection. The dissection then is continued inferior and posterior to the incision. After a sufficient area of chest wall has been mobilized, a chest retractor may be inserted.
S-shaped posterolateral thoracotomy incision extending to the costal margin provides excellent exposure.
A. Sagittal section of right lung and pleura. B. Pleurectomy is begun between the endothoracic fascia and the parietal pleura using blunt dissection. The plane is developed cephalad.
Care must be taken in dissecting around the subclavian vessels in the lung apex.
The pleura now can be mobilized from the mediastinum. Once the upper portion of the lung is completely mobilized from the chest wall, the superior and posterior hilar structures are well exposed. On the left side, the esophagus and aorta must be identified and the dissection around them should be undertaken with care. On the right side, the superior vena cava must be dissected away from the specimen gently. The dissection then continues to the posterior aspect of the pericardium. A plane between the mediastinal pleura and the pericardium is sometimes present. If it is not, the pericardium needs to be resected en bloc at a later stage of the operation with subsequent reconstruction. The dissection then is carried toward the posterior diaphragmatic sulcus. If superficial involvement of the diaphragm is found, a partial-thickness resection can be performed. The plane between the tumor and the uninvolved diaphragm can be entered, and the dissection is initiated at the posterior costophrenic angle and carried anteriorly. This is facilitated by strong retraction on the pleura away from the diaphragm. In many patients, deeper involvement of the diaphragm mandates a full-thickness resection of a portion of the muscle. The deep border of the diaphragm then must be dissected from the peritoneum. Care should be taken to avoid entering the abdomen because tumor seeding into the peritoneal cavity is a concern. This is often unavoidable, especially around the central tendon, and any defect in the peritoneum should be closed immediately. The specimen then is mobilized en bloc back toward the pericardium medially. If resection of the pericardium is required, it is delayed until the tumor is mobilized as much as possible owing to the accompanying arrhythmias from manipulation. The pericardium is opened gradually, and traction sutures are placed on the nonspecimen edge to maintain the position of the heart and to prevent retraction of the pericardium into the opposite hemithorax (Fig. 121-4).
The parietal pleura has been stripped from the chest wall, and the lung is ready for decortication. Note the exposed hilar structures, traction sutures on the pericardium, closed peritoneal defect, and resection of involved diaphragm.
Once the dissection is completed to the hilar structures, the parietal pleura is opened, and the pleural envelope is entered and decortication of the visceral pleura is performed. This is, in some respects, the most technically demanding and tedious component of the operation. Decortication must be performed with care into the fissures because they are often substantially involved with disease (Fig. 121-5). During the decortication, deflation of the lung will minimize blood loss, and inflation will allow better visualization of the plane between the tumor and the visceral pleura or lung parenchyma. Communication with the anesthesiologist about the amount of blood loss is important because most patients require intraoperative transfusion.
Decortication must be performed meticulously in the fissures to remove all gross diseases.
Lymph node dissection should be carried out, and specimens should be labeled and sent separately to the pathologist. The subcarinal lymph nodes should be resected as well as the paratracheal lymph nodes on the right and the aortopulmonary lymph nodes on the left.
Once the gross tumor is removed and the specimen is delivered, reconstruction of the pericardium and diaphragm, if required, is performed. If the diaphragm is largely intact, it can be closed primarily by plication to prevent upward movement and subsequent compression atelectasis of the lower lobe. On the right side, reconstruction of the diaphragm is performed with a double layer of Dexon (United States Surgical, Syneture Division, Norwalk, CT) mesh because the liver prevents herniation of intra-abdominal contents. On the left, 2-mm-thickness Gore-Tex (W.L. Gore and Associates, Flagstaff, AZ) is used because thicker, nonabsorbable material is required to prevent herniation. The prosthesis is secured laterally by placing sutures around the ribs. Posteriorly, it is sutured to the crus or tacked to the prevertebral fascia. The medial aspect is sewn to the remaining edge of the diaphragm at its confluence with the pericardium. The diaphragmatic prosthesis should be made absolutely taut to prevent upward motion of the abdominal contents and subsequent atelectasis of the lower lobe. If the pericardium was resected, it is reconstructed with a single layer of Dexon mesh.
Attention is now turned to obtaining hemostasis. An argon beam electrocoagulator may be used to help control diffuse bleeding from the chest wall. Three chest tubes are placed anteriorly and posteriorly into the apex, and a right-angle tube is placed along the diaphragm. This should permit control of the substantial air leaks that are anticipated and should permit full expansion of the lung. The air leaks tend to resolve after 72 hours if the lung is fully expanded.
During the dissection, certain areas of particular concern warrant special mention. The subclavian vessels can be injured by traction during the blunt dissection of the apex. On the right, care needs to be taken while dissecting the mediastinal pleura from the superior vena cava. On the left, the plane between the tumor and the adventitia of the aorta, the origins of the intercostal vessels, and the esophagus all should be identified. If the diaphragm is largely left intact and reconstruction is not undertaken, plication is often helpful to prevent elevation and paradoxical motion of the diaphragm and atelectasis of the lower lobe.
Although prior talc pleurodesis is not an absolute contraindication to the operation, it does increase the likelihood of substantial blood loss and air leak.
Insuring adequate lung expansion at the end of the operation is critical. As there will be numerous breaches of lung parenchyma, a substantial degree of postoperative air leak is expected. Failure to re-expand the lung will then result in a fixed space defect in the setting of prolonged air leak—a recipe for substantial postoperative problems including empyema.
As there is usually a substantial degree of air leak in the immediate postoperative period, drains are placed on suction, if tolerated, overnight. As long as there is good lung expansion, the air leak settles down in the first few postoperative days and suction can usually be removed on postoperative day one.
Postoperative complications are similar to most thoracic procedures. Atrial fibrillation rates as high as 17% have been reported.17 Prolonged air leak due to the inadequate lung expansion and subsequent empyema is particularly difficult to manage especially if a prosthesis is in situ (diaphragmatic or pericardial repair) and would mandate reoperation.
P/D is generally well tolerated, with a mortality rate limited to approximately 1% to 2% when performed at high-volume centers. The most common complication is prolonged air leak, which occurs in 10% of patients. Hemorrhage, pneumonia, and empyema are less common complications.18The results of studies examining P/D alone are summarized in Table 121-1. Median survivals range from 9 to 20 months in the literature.
Table 121-1Studies of Pleurectomy Decortication Alone ||Download (.pdf) Table 121-1Studies of Pleurectomy Decortication Alone
|SOURCE ||YEAR ||PATIENTS (N) ||MEDIAN SURVIVAL (MOS) ||MORTALITY (%) |
|Chahinian et al.19 ||1982 || 30 ||13 ||0 |
|Brenner et al.20 ||1982 || 69 ||15 ||N/A |
|Law et al.21 ||1984 || 28 ||20 ||11 |
|Chailleux et al.22 ||1988 || 29 ||14 ||N/A |
|Ruffie et al.23 ||1989 || 63 ||10 ||0 |
|Brancatisano et al.24 ||1991 || 45 ||16 ||2 |
|Rusch et al.25 ||1991 || 26 ||10 ||N/A |
|Allen et al.26 ||1994 || 56 || 9 ||5 |
|Soysal et al.27 ||1997 ||100 ||17 ||1 |
|Ceresoli et al.28 ||2001 || 38 ||13 ||N/A |
The technical challenge of separating tumor and visceral pleura from the lung parenchyma may result in suboptimal cytoreduction. This is reflected in the observation that the most common site of recurrence is the ipsilateral hemithorax.29
Since the results of surgery alone are poor, most recent studies have combined P/D with some form or combination of adjuvant therapy. These have included external radiation, brachytherapy, systemic chemotherapy, intrapleural chemotherapy, and photodynamic therapy (PDT). It is important to note that these studies are almost uniformly observational in nature. When comparisons are performed, they are by and large across inhomogeneous groups, thereby limiting the conclusions that can be drawn about efficacy.
Studies that have used various forms of radiation therapy are summarized in Table 121-2. The earliest experience with combined therapy for MPM was reported by McCormack et al.8 at the Memorial Sloan-Kettering Cancer Center. The combination of P/D with external radiation and systemic chemotherapy in 18 patients with epithelial mesothelioma produced a median survival of 16 months. In the subsequent 33 patients, brachytherapy was added, and the median survival was 21 months.
Table 121-2Studies of Pleurectomy with Radiation Therapy ||Download (.pdf) Table 121-2Studies of Pleurectomy with Radiation Therapy
|Source ||Therapy ||Year ||Patients (n) ||Median Survival (mos) ||Mortality (%) |
|McCormack et al.8 ||S, R, C ||1982 || 18 ||16 ||2 |
|Alberts et al.30 ||S, R, B ||1988 || 26 ||11 ||N/A |
|Mattson et al.31 ||S, R, C ||1992 ||100 ||8 ||N/A |
|Lee et al.32 ||S, R, ± C ||2002 || 26 ||18 |
In another study at the Memorial Sloan-Kettering Cancer Center, brachytherapy was used in patients who had gross residual disease after P/D, followed by postoperative external beam radiation therapy (median dose of 4200 cGy). Local failure or disease progression occurred in 63% of patients, and median survival was 13 months.33,34 Alberts et al. evaluated 262 patients with MPM, 26 of whom had P/D followed by radiation and chemotherapy.30 The median survival for the whole group was 9.6 months and for the subset undergoing surgery survival was 10.9 months, which was not a statistically significant difference.
A Finnish study consisting of 100 patients evaluated five different adjuvant radiation schedules and chemotherapy regimens following pleurectomy.31 The median survival was 8 months, and the 2-year survival was 20%. The authors found no difference among the groups.
In a more recent study, Lee et al.32 from the University of California (UCLA), San Francisco reported their experience with 32 patients who had undergone P/D with intraoperative radiotherapy followed by external beam radiation. Some patients received chemotherapy as well. The median survival of the 26 patients who underwent the planned treatment was 18.1 months. Not included in the analysis are the three patients who had unresectable disease, one with recurrent disease and two who had early postoperative deaths.
In patients deemed unfit for EPP who undergo P/D at the Memorial Sloan-Kettering Cancer Center, our current practice is to give adjuvant external beam radiation therapy.35
P/D with Intrapleural and/or Systemic Chemotherapy
Studies that have been conducted using intrapleural or systemic chemotherapy are summarized in Table 121-3. Rusch et al.36 from the Memorial Sloan-Kettering Cancer Center evaluated P/D followed by intrapleural cisplatin and mitomycin followed by systemic cisplatin and mitomycin given 3 to 5 weeks postoperatively. Twenty-eight patients underwent P/D and received intrapleural chemotherapy. There was one postoperative death, and two patients developed grade 4 nephrotoxicity. The median survival was 18 months, and significant morbidity was present in 53% of patients. Local failure was high, with 16 local relapses in 27 patients. The authors were concerned about the potential for serious toxicity.
Table 121-3Studies of Pleurectomy with Chemotherapy ||Download (.pdf) Table 121-3Studies of Pleurectomy with Chemotherapy
|Source ||Therapy ||Year ||Patients (n) ||Median Survival (mos) ||Mortality (%) |
|Rusch et al.36 ||S, I, C ||1994 ||28 ||18 ||4 |
|Rice et al.37 ||S, I, C ||1994 || 9 ||13 ||5 |
|Lee et al.32 ||S, I ||1995 ||15 ||12 ||0 |
|Colleoni et al.38 ||S, I, C ||1996 ||20 ||12 ||0 |
|Hasturk et al.39 ||S, C ||1996 ||20 ||12 ||N/A |
|Ceresoli et al.28 ||S, C ||2001 ||16 ||14 ||N/A |
|Sugarbaker et al.43 ||S, I ||2003 ||44 || 9 ||11 |
Rice et al.37 studied 19 stage I MPM patients who had EPP (n = 10) or P/D (n = 9) followed by intrapleural administration of cisplatin and mitomycin. The median survival was 13 months, and the treatment related mortality rate was 5%.
A group from UCLA reported their results on 15 patients who underwent P/D followed by intrapleural cisplatin and cytarabine.32 The median survival was 11.5 months with no treatment-related mortality. A similar study was reported by an Italian group that added systemic chemotherapy to the same regimen of P/D and intrapleural chemotherapy.38 The 20 patients in this study also had an 11.5-month median survival.
A group from Turkey reported on 20 patients who had P/D followed by systemic chemotherapy consisting of cisplatin, mitomycin, and α-interferon immunotherapy.39 The median survival was 12 months, and the regimen was well tolerated.
In a more recent study, Ceresoli et at.28 from Italy retrospectively reviewed their experience with MPM and noted that the 16 patients who had P/D followed by chemotherapy did better than patients who received P/D or chemotherapy alone. The median survivals were 14 months in the combined treatment group, 12 months in the surgery alone group, and 8 months in the chemotherapy alone group. In univariate analysis, this treatment modality had independent prognostic value.
The search for effective therapies to achieve local control in MPM has sparked interest in a number of areas. Hyperthermia has been investigated in combination with surgery. Carry et al.40 reported the results in three patients with MPM who were given hyperthermic intrapleural mitomycin at 42.6°C for 60 minutes following P/D. The technique was deemed safe.
In a study of similar design, Ratto et al.41 gave hyperthermic cisplatin (41.5°C) for 60 minutes following P/D (three patients) or EPP (four patients). Patients also received 55 Gy to chest wall incisions. There was no death or toxicity, and the treatment was well tolerated. An interesting finding was that systemic cisplatin levels were significantly higher in the group that had P/D, indicating that the remaining lung plays an important role in the absorption of intrapleural cisplatin.
A group from the Netherlands studied the use of intrapleural hyperthermic (40°C–41°C) cisplatin and doxorubicin after P/D.42 There was considerable toxicity, with morbidity reported in 47% of patients. The 11 patients had a median survival of 8 months. This group used the same protocol for advanced thymoma patients.
In a dose-escalation study, Sugarbaker et al.43 performed P/D on 44 patients followed by intraperitoneal and ipsilateral hemithoracic lavage with cisplatin at 42°C. They reported a postoperative mortality rate of 11% and median survival of 9 months.
P/D and Photodynamic Therapy
Studies of PDT, a new modality used to enhance local control, are shown in Table 121-4. A photosensitizer is administered systemically, and then target areas are illuminated with laser to affect cell kill. Moskal et al.44 from Roswell Park reported their series of 40 patients who had been treated with EPP (n = 7), P/D (n = 28), or P/D with lobectomy (n = 5) followed by PDT.45 The mortality rate was 6.5%, with two of the deaths in the EPP group. Serious complications arose in 48.3% of patients. The median survival was 15 months.
Table 121-4Studies of Pleurectomy with Photodynamic Therapy ||Download (.pdf) Table 121-4Studies of Pleurectomy with Photodynamic Therapy
|SOURCE ||THERAPY ||YEAR ||PATIENTS (N) ||MEDIAN SURVIVAL (MOS) ||MORTALITY (%) |
|Pass et al.46 ||S, P, C, I ||1997 ||25 ||14 ||4 |
|Moskal et al.44 ||S, P ||1998 ||40 ||15 ||7 |
Pass et al.46 at the National Institutes of Health performed a randomized study of surgical resection, postoperative cisplatin, interferon, and tamoxifen with or without PDT. Twenty-five patients received PDT (11 P/D and 14 EPP), and 23 patients did not (12 P/D and 11 EPP). The groups were similar, and no survival difference was noted. Median survivals in the PDT group and non-PDT group were 14.4 and 14.1 months, respectively. The conclusion was that there was no value of first-generation PDT when added to multimodality therapy. Further studies with newer photosensitizers and increasing light doses are ongoing.